Plasma Science and Technology, Vol.17, No.11, Nov. 2015

Microanalysis of Multi-Element in Juncus effusus L. by LIBSTechnique∗

LIU Xiaona (刘晓娜)1, HUANG Jianmei (黄建梅)1, WU Zhisheng (吴志生)1,ZHANG Qiao (张乔)1, SHI Xinyuan (史新元)1,2, ZHAO Na (赵娜)1,JIA Shuaiyun (贾帅芸)1, QIAO Yanjiang (乔延江)1,2

1College of Chinese Medicine, Beijing University of Chinese Medicine, Beijing 100102, China2Research Center of TCM-information Engineering, State Administration of TraditionalChinese Medicine of the People’s Republic of China, Beijing 100102, China

Abstract Laser-induced breakdown spectroscopy (LIBS) was used to decipher the unique

multi-elemental characteristics of Juncus effusus L. The spectral fingerprints of Juncus effusus L.

were established based on elemental microanalysis via LIBS. Microanalysis and multimode sam-

pling methodologies were designed in this study. The relative standard deviation (RSD) approach

was performed to optimize the multi-shot measurements. Taking advantage of the capability with

no or minimal sample pre-treatment of LIBS, a thermodynamic chart of four elements (Mg, Ca,

Ba, and Na) was created from twelve collection regions. The diagram of elemental distribution on a

micro-scale was generated to explore the nature of Juncus effusus L. by LIBS. The results demon-

strated that LIBS is a promising technique for rapid elemental microanalysis of heterogeneous

samples.

Keywords: LIBS, Juncus effusus L., RSD approach, thermodynamic chart, elementalmicroanalysis, micro-scale

PACS: 87.64.−t

DOI: 10.1088/1009-0630/17/11/02

(Some figures may appear in colour only in the online journal)

1 Introduction

The dried stem of the whole aerial part of Juncus ef-fusus L. is traditionally used in China to treat fidgetingand insomnia, a collection of symptoms related to anx-iety [1]. Juncus effusus L. is a heterogeneous medicalplant primarily comprised of macromolecular cellulose,hemicelluloses, lignin, and low molecular-weight sub-stances, i.e. extractives, and minerals [1,2].

Most fingerprint studies of Chinese Materia Med-ica (CMM) have focused on organic composition ratherthan examining the inorganic elements [3−5]. However,the mineral compositions have a great impact on thequality of the herb. Meanwhile, numerous plants canbe used as bio-indicators of the environment [6]. Themineral composition of plants is related to external pa-rameters such as soil, fertilizer, contamination, and dis-turbances involving human activity. Juncus effusus L.,a wetland plant, is an ideal bio-indicator. Wu et al.reported that Juncus effusus L. was used as a wetlandfilter to treat drinking water [7]. Therefore, it is sig-nificant to rapidly and accurately monitor the mineralelement of Juncus effusus L.

Advances in chemical science have led to an increas-ing demand of microanalytical techniques that allow a

spatially resolved analysis of solid samples without anextensive sample pretreatment. The conventional beamand particle techniques, such as electron probe and sec-ondary ionization mass spectrometry, can analyze spa-tial features on a micro-scale with high sensitivity [8].However, they need the necessary experimental condi-tions, i.e. high vacuum, laborious sample pretreatment,or rigid matrix-related interference corrections. On theother hand, fast information on spatially resolved el-emental distributions on a micro-scale is required inmany production processes.

Laser-induced breakdown spectroscopy (LIBS), anadvanced elemental microanalysis technique, is a pow-erful tool to investigate structure, chemical, or mineralcomponents on the micro-scale [9]. LIBS can detectdiverse materials, including solids, liquids, gases andaerosols with no or minimal sample pretreatment [10].It relies on plasma microanalysis [11,12]. Plasma is in-duced by a high energetic laser pulse. Sufficient irradi-ance enables the generation of ionized plasma includ-ing atoms and ions. Subsequent detection of the spe-cific spectral emission reveals the information about el-emental compositions of samples LIBS provides fast, insitu, and even remote multi-element analysis that hastriggered some additional interest in the application of

∗supported by National Natural Science Foundation of China (No. 81303218), Beijing Municipal Government for the UniversityAffiliated with the Party Central Committee, and Doctoral Fund of Ministry of Education of China (No. 20130013120006)

904

LIU Xiaona et al.: Microanalysis of Multi-Element in Juncus effusus L. by LIBS Technique

LIBS [13−15].Recently, LIBS has drawn interest for application of

plant materials, notably for a chemical sensor tech-nique to analyze both macro- and micronutrients inplants [16,17]. Santos et al. reported the LIBS techniquein the application of plant materials [18]. Krizkova et al.studied the response of sunflower plants to stress in-duced by silver (I) ions as well as the basic physiologi-cal parameters by LIBS [6]. Galiova et al. used LIBS toinvestigate heavy-metal accumulation in selected plantsamples [19]. Liu et al. explored the rapid elemen-tal analysis and provenance study of a medicinal herb(Blumea balsamifera DC) using LIBS [20].

LIBS also provides information about organic ele-ments, i.e. carbon (C), hydrogen (H), nitrogen (N),and oxygen (O) together with molecular bands (C-Nand C-C), which are difficult to detect by other con-ventional techniques. Based on multivariate analysis,the LIBS technique has shown good classification per-formance. Wang et al. studied the physical mecha-nism of laser-induced plasma of an organic explosiveand demonstrated that LIBS coupled with the chemo-metric techniques had the capacity to discriminate anorganic explosive from plastics [21].

The aim of this study is focused on the microanalysisof Juncus effusus L. by LIBS. Robust instrumentationand user-friendly methodologies were designed to studythe nature of heterogeneous samples on a micro-scale.Specifically, the relative standard deviation (RSD) ap-proach and elemental thermodynamic chart were ap-plied to process the spectral data and visualize the dis-tribution of the main elements.

2 Materials and methods

2.1 Experimental setup

A commercial LIBS system (TSI, ChemRevealTM-3764, USA) equipped with a Q-switched Nd:YAG laserwas employed to collect spectra. Experiments were per-formed at ambient air with a laser energy about 340 mJper pulse, repetition rate of 2 Hz, and pulse durationof 3-5 ns at 1064 nm. The spectrum covers a contin-uous wavelength from 170 nm to 985 nm. A detectordelay of 1 µs and a fixed spectrometer integration timeof 1 ms were used. These values resulted from an opti-mization study carried out for the detection of 21 stan-dard reference materials (SRM), including food, clay,sludge, steelmaking alloy, and geochemical and agricul-tural materials [22]. The optimization study was alsoused to analyze carbon in coal [23]. The laser beampassed through a pierced parabolic mirror and was fo-cused vertically onto the sample. The plasma was col-lected by a 50 mm focal length lens and optical fibers.It was then fed into the spectrometer and recorded by 7-channel-charge-coupled device (CCD) camera. A three-dimensional (3D) translation stage with stepper motorswas used to ensure accurate movement of the sample.The LIBS data system was supported by ChemRevealLIBS software.

2.2 Materials

Plant materials were collected from different regionsof China (shown in Table 1). The samples were ei-ther the dried stem or medulla of Juncus effuses L.,which were confirmed to be Juncus effuses L. by Dr.Jian-mei Huang (a principle investigator of the Insti-tute of Chinese medicine, Beijing University of Chi-nese Medicine). The samples were designated S1-S12.The voucher specimens (No. GT110 - No. GT121) werekept at the Herbarium of Beijing University of ChineseMedicine.

Table 1. A summary of the tested samples

Sample codes Collection regions

S1 Guizhou

S2 Anhui

S3 Bozhou, Anhui

S4 Bozhou, Anhui

S5 Shizhen large pharmacy

S6 Jiangxi

S7 Heilongjiang

S8 Xi’an, Shanxi

S9 Guangzhou

S10 Laifeng xian, Hubei

S11 Fuzhou, Fujian

S12 Hubei

2.3 Data acquisition and microanalysis

A schematic diagram of LIBS experiment is shownin Fig. 1. Spectra were obtained on the micro-scaleon the samples’ surface. For each sample, ten micro-locations were randomly selected along the sample andeach micro-location consisted of nine laser pulses (ma-trix, 3×3 (100 µm × 100 µm)).

2.4 Data analysis

Data analysis was performed under the Windows7 Professional operating system using Matlab R2014a(Mathworks, Inc., Natick, MA).

2.4.1 The relative standard deviation approach

The sampling source is one of the fundamental pointsof LIBS toward accurate analysis. Matrix effects are ofcritical importance in several analytical spectroscopytechniques, and LIBS is no exception to this [23]. Anapproach of LIBS spectra described in the case of ahighly heterogeneous system, named the relative stan-dard deviation (RSD) approach [12], was applied to op-timize the number of laser shots during sampling. RSDapproach is based on a single-shot and individual spec-trum. As stated in Ref. [12], the conventional spectracould be acquired by plotting the signal versus wave-length, while the “RSD spectra” were obtained by plot-ting RSD versus wavelength.

905

Plasma Science and Technology, Vol.17, No.11, Nov. 2015

Fig.1 A schematic diagram of data acquisition on a micro-scale during LIBS experiment

The observed profile of the “RSD spectra” depends onthe source of the limiting noise, the relative intensityof the signal, and their correlation. Thus, “RSD spec-tra” can be used to choose the optimal sampling mode.With a little difference, “RSD spectra” were used toassess the multi-shot acquisition in this set of investi-gations.

2.4.2 Elemental thermodynamic chart

An elemental thermodynamic chart was used to ex-plore the distribution of different elements in plants andto further study the dominant signal in “RSD spec-tra”. In an elemental thermodynamic chart, differentcolors demonstrate different intensities of spectral lines.For example, in an elemental thermodynamic chart redrepresents high intensity of emission lines while yellowrepresents low intensity of emission lines. Thus, thedistribution of elements could be visualized directly.

3 Results and discussion

3.1 Elemental identification of Juncuseffuses L.

The identification of spectral lines was performed byusing the National Institute of Standards and Tech-nology (NIST) database and relative references [24−26].The representative LIBS spectra are shown in Fig. 2(2).

According to the results of qualitative detection bythe LIBS technique, 33 emission lines of 12 elementsand molecular bands were identified to establish a spec-tral “fingerprint”. The alkali metal and alkaline earthmetals (Ca, K, Ba, Na) were clearly detectable becausethey dominate the LIBS spectra. Magnesium (Mg) wasalso detected. The light organic elements such as car-bon (C), hydrogen (H), nitrogen (N) and oxygen (O)together with C-N molecular bands were also simulta-neously monitored. Table 2 lists the elements observedin the spectra and their corresponding emission wave-lengths.

Fig.2 (1) The sampling sketch map of ten micro-locations for each sample, (2) the representative LIBS spectra of Juncus

effuses L., (3) “RSD spectra”: (a) n = 1 × 9 (one micro-location), (b) n = 5 × 9 (five separate micro-locations), and (c)

n = 10× 9 (ten separate micro-locations)

906

LIU Xiaona et al.: Microanalysis of Multi-Element in Juncus effusus L. by LIBS Technique

Table 2. The observed LIBS emission lines of Juncus effusus L. samples

Element Wavelength (nm) Element Wavelength (nm)

C 192.77, 247.725 Na 588.952, 589.554

Mg 279.418, 280.123 H 656.315, 777.492

Ca 393.375, 396.816 Li 670.754

C-N 383.522, 386.105, 387.08, 388.296 K 766.52, 769.96

Al 394.417, 396.097 O 777.212, 777.492

Ba 455.358, 493.388, 553.52 N 715.709, 744.306, 746.918, 843.762,

Mg 517.245, 518.316 869.367, 870.256, 870.947

3.2 “RSD spectra” of Juncus effuses L.

Fig. 2(1) shows the sampling characteristic (such asone micro-location, five and ten micro-locations) andrepresentative LIBS spectra of Juncus effuses L. “RSD-spectra” at one micro-location (nine pulses measure-ment) and separate micro-locations (five and ten) persample are shown in Fig. 2(2).

In the “RSD spectra”, the dominant noise wouldbe the background, which indicates the measurementprecision during the plasma evolution. Moreover, thedominant signal demonstrates the mineralogical vari-ability among heterogeneous samples. In “RSD spec-tra” of Juncus effuses L., the dominant signals wereMg 279.418 nm, Ca 393.375 nm, Ba 455.358 nm, andNa 588.952 nm, 589.554 nm, etc. Further analysis willbe given in the following section.

Fig. 3 is a detailed graph of Fig. 2(3) in a wavelengthof 660-740 nm. In general, the values of RSD decreasedwith the increasing number of laser pulses. Firstly, thevalues of RSD at one micro-location (n=9) were lessthan 0.08. Secondly, the values of RSD at five micro-locations (n=9×5) were less than 0.06. Finally, at tenseparate micro-locations (n=9×10) the values of RSDspectra were less than 0.05. Therefore, a total of 90shot measurements per sample at ten separate micro-locations were better for the spectroscopic analysis.

Fig.3 The detailed graph of Fig. 2(3) in wavelength of

660-740 nm

3.3 Elemental distribution of Juncuseffuses L.

The thermodynamic chart of four elements (Mg, Ca,Ba, and Na) obtained at different micro-locations acrossthe medulla of Juncus effuses L. was used to explore theelemental distribution (as shown in Fig. 4).

For the 12 samples, the color of four elements was dif-ferent. Moreover, the color of four elements at differentmicro-locations was distinct, especially for Ca and Ba.In Fig. 4(a), the image of Mg was dark red or light red

Fig.4 Intensity dependent thermodynamic chart of four elements at ten different micro-locations (“ml”) on the micro-surface of stem samples from twelve collection regions. (a) Mg 279.418 nm, (b) Ca 393.375 nm, (c) Ba 455.358 nm, and (d)Na 588.952 nm. The color legend (bar on the right) refers to the intensity of spectral lines

907

Plasma Science and Technology, Vol.17, No.11, Nov. 2015

overall, whereas the color was alternating between redand yellow in S9 and S12 at different micro-locations.This meant that the distribution of Mg was uneven inS9 and S12.

In Fig. 4(b), the red color was alternating in thechart, which demonstrated that the distribution of Cawas heterogeneous in numerous samples. Moreover, thecolor of S3 was wine in the diagram illustrated the ho-mogenous distribution of Ca. Clearly, the intensity ofBa spectral line was homogeneous in all of the samplesexcept S7 (as shown in Fig. 4(c)) because the image ofBa spectral line was light yellow at “ml-1” of S7.

Additionally, the intensity of Na spectral linechanged greatly in numerous samples, especially in S4(as shown in Fig. 4(d)). The diagram of Na showedlight yellow at the “ml-8”, “ml-9” and “ml-10” of S4,which meant that the intensity of Na spectral line washigher than that of other micro-locations. Generally,the distribution of Na was heterogeneous at differentmicro-locations due to the large variation of color. Theintensity of the four elements was low and relativelysteady in S3 because the diagram of S3 was red in color.

These results of intensity dependent thermodynamicchart of four elements elucidated the mineralogical vari-ability among the samples. The intensity (or the quan-tity) of elements was distinct in the heterogeneous sam-ples. Therefore, the multimode sampling methodologyof microanalysis was significant for the heterogeneoussamples.

4 Conclusions

In this investigation, LIBS was found to be a valu-able tool for the analysis of Juncus effusus L. samples.Specifically, microanalysis and RSD approach were usedto explore a multimode sampling method for heteroge-neous samples. Furthermore, the thermodynamic chartof four elements (Mg, Ca, Ba and Na) successfully char-acterized the mineralogical distribution of Juncus ef-fusus L. samples on a micro-scale.

Due to a popular Chinese herb and an ideal bio-indicator, the microanalysis of Juncus effusus L. re-search on the spectral “fingerprint” is significant. Thesimilar algorithms can be triggered in the microanalysisof LIBS within plant materials. CMM has its own char-acteristic and it is the treasure of China. The ChineseLIBS community is one of the most dynamically devel-oping communities in the world [27]. LIBS has potentialin a CMM based on microanalysis.

Acknowledgment

The authors are very grateful to Rock Liu and TSI,Inc. (St Paul, MN, USA) for technical support and thefree access of instrument.

References

1 Chinese Pharmacopoeia Commission. 2010, Pharma-copeia of People’s Republic of China, 1st ed. China

Medical Science Press, Beijing (in Chinese)

2 Liao Youjiao, Zhai Haifeng, Zhang Bing, et al. 2011,Planta Medicine, 77: 416

3 Xie Peishan, Chen Sibao, Liang Yizeng, et al. 2006,Journal of Chromatography A, 1112: 171

4 Chen Dan, Zhao Sandyshuo, Leung Kelvinszeyin.2009, Journal of Separation Science, 32: 2892

5 Zhang Qingfeng, Cheung Honyeung, Zeng Lingbin.2013, Journal of Natural Medicines, 67: 207

6 Krizkova Sona, Ryant Pavel, Krystofova Olga, et al.2008, Sensors, 8: 445

7 Wu Min, Li Qingyun, Tang Xianqiang, et al. 2014, In-ternational Journal of Environmental and AnalyticalChemistry, 94: 618

8 Alvira F C, Ramirez Rozzi F, Bilmes G M. 2010, Ap-plied Spectroscopy, 64: 313

9 Godwal Y, Taschuk M T, Lui S L, et al. 2008, Laserand Particle Beams, 26: 95

10 Pathak Ashok Kumar, Kumar Rohit, Singh Vivek Ku-mar, et al. 2012, Applied Spectroscopy Reviews, 47: 14

11 Hahn David, Nicolo Omenetto. 2010, Applied Spec-troscopy, 64: 335

12 Hahn David, Nicolo Omenetto. 2012, Applied Spec-troscopy, 66: 347

13 Pinon V Mateo M P, Nicolas G. 2013, Applied Spec-troscopy Reviews, 48: 357

14 Cai Yue, Chu Pochun, Ho Sutkam, et al. 2012, Fron-tiers of Physics, 7: 670

15 Dong Fengzhong, Chen Xinglong, Wang Qi, et al.2012, Frontiers of Physics, 7: 679

16 Trevizan Lilian Cristina, Santos Jr. Dario, Samad Ri-cardo Elgul, et al. 2009, Spectrochimica Acta Part B,64: 369

17 Trevizan Lilian Cristina, Santos Jr. Dario, Samad Ri-cardo Elgul, et al. 2008, Spectrochimica Acta Part B,63: 1151

18 Santos Jr. Dario, Nunes Lidiane Cristina, CarvalhoGabriel Gustinelli Arantes de, et al. 2012, Spec-trochimica Acta Part B, 71-72: 3

19 Galiova M, Kaiser J, Novotny K, et al. 2008, AppliedPhysics A: Materials Science and Processing, 93: 917

20 Liu Xiaona, Zhang Qiao, Wu Zhisheng, et al. 2015,Sensors, 15: 642

21 Wang Qianqian, Liu Kai, Zhao Hua, et al. 2012, Fron-tiers of Physics, 7: 701

22 Choi Soojin, Lee Kangjae, Yoh Jack J. 2013, AppliedPhysics B, 113: 379

23 Wang Zhe, Hou Zongyu, Lui Siulung, et al. 2012, Op-tics Express, 20: A1011

24 Li Xiongwei, Wang Zhe, Lui Siulung, et al. 2013, Spec-trochimica Acta Part B, 88: 180

25 Alvey Daniel C, Morton Kenneth, Harmon Russell S,et al. 2010, Applied Optics, 49: C168

26 Corsi Michela, Cristoforetti Gabriele, HidalgoMontserrat, et al. 2006, Applied Geochemistry,21: 748

27 Wang Zhe, Yuan Tingbi, Hou Zongyu, et al. 2014,

Frontiers of Physics, 9: 419

(Manuscript received 25 April 2015)(Manuscript accepted 29 June 2015)E-mail address of corresponding authorQIAO Yanjiang: yjqiao@263.net

908